Question: How can I filter/extract called peaks based on their peak height?
0
gravatar for jiekwo1
11 weeks ago by
jiekwo10
jiekwo10 wrote:

Hello everyone,

So I am working with ATAC-seq data and wondering if there is a way to extract peaks that are above a certain threshold. For example, I want to extract peaks that are above the red line shown in the picture. Is this something I need to do at the time of peak calling? or can it be done after peak calling?

I am very new to ATAC-seq field so it's difficult to phrase what I am looking for into the right sentence. so please bear with me.

For your information, I just have bed and bigwig files.

If anyone could walk me through how to do that, I would greatly appreciate it!

All the best,

atac-seq-aboveline

ADD COMMENTlink modified 11 weeks ago • written 11 weeks ago by jiekwo10
1

Technically you could do this based on some custom scripting that first extract the summit coverage and then subsets the peaks but that is cumbersome. I would rather filter for significance as peak height and width alone is influenced by a number of factors including technical ones such as mappability and GC content of the underlying DNA sequence. If you find that macs2 produces many small non-sensish peaks then yes I agree, that is why I abandoned it for ATAC-seq. I mean, it was developed for ChIP-seq where you had a lot more noise compared to the relatively clean ATAC-seq data. I like the Genrich peak caller, see https://github.com/jsh58/Genrich. You can set parameter for minimal peak length, AUC, and significance and it even can take replicated samples.

ADD REPLYlink modified 11 weeks ago • written 11 weeks ago by ATpoint44k

thanks for your reply! I am going to try Genrich.

ADD REPLYlink written 11 weeks ago by jiekwo10
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2215 users visited in the last hour
_